Development of soft ionization methods, like Electrospray (ESI) and Matrix Assisted Laser Desorption/Ionization (MALDI) have extended the field of mass spectrometric analysis to wide class of labile compounds—such as peptides, nucleotides, proteins, and lipids—and have triggered the development of wide range of biological and medical applications. The methods are known to be limited to compounds which readily produce ions in liquid, such as ionic salts, and to polar compounds, readily producing protonated MH+ and deprotonated ions (M−H)−. The range of soft ionized compounds was extended to semi-polar compounds with introduction of Atmospheric Pressure Chemical Ionization (APCI) and Atmospheric Pressure Photo Ionization methods (APPI) methods. One shortcoming, however, is that these soft ionizing methods are not fully quantitative because the nature of the analyzed compounds define and vary both the ionization efficiency and the gas-phase stability against competitive ion molecular reactions.
On the other pole, the truly quantitative method of electron impact ionization (EI), wherein compound vapors are ionized by electron bombardment, has long existed. The ionization efficiency stays constant in wide range of analyte concentration, usually measured as sample load through a gas chromatograph (GC). Typically, linearity is sustained from a limit of detection (LOD) being as low as ten femtograms (10 fg) in most sensitive instruments and up to ten nanograms (10 ng) load range (i.e. at least within six orders of dynamic range). The ionization efficiency (i.e. response versus load) is mostly non-dependent on molecule nature, and stays independent on other coeluting compounds and matrix. This allows the EI method to be uniform across chemical classes and truly quantitative. The EI method, however, is limited to semi-volatile compounds, and it is not soft is not soft because it produces extensive fragmentation.
In addition to coupling with an EI method source, GC separation has been coupled to alternative and notably softer ionization methods—such as chemical ionization (CI) and field ionization (FI)—which provide more intensive molecular peak. The CI technique, however, is also prone to matrix and mutual interference effects. With this ionization method, both ionization efficiency and spectra content strongly depend on instrumental parameters. Thus, the CI technique is not considered to be fully soft, truly quantitative, or capable of providing library spectra. And the CI technique is also considered “dirty” due to the rapid contamination of the ion source. The FI method is frequently regarded as a soft ionization method; however, it is tricky, unstable, and insensitive with typical detection limit of only around one hundred picograms (100 pg). For this reason, the FI method has not been widely adopted.
Photo ionization (PI) and photo-chemical ionization (APPI) methods are much softer compared to EI, though still produces fragments for highly fragile compounds. Schlag describes in U.S. Pat. No. 4,570,066, which is fully incorporated herein by reference, that multi-photon ionization for laser desorbed nucleotides and short peptides, along with their cooling by a supersonic jet with subsequent multi-photon resonance ionization, which appeared to be moderately soft. The method was not widely adopted due to selective ionization, insufficient softness, and a limited class of analyzed compounds.
Glow discharge has been long employed in mass spectrometry for elemental and organic analysis, such as in F. W. Aston, MASS SPECTRA AND ISOTOPES, 2nd edition, Longman Green, New York, 1942, which is fully incorporated herein by reference. In Hunt et. al, Anal. Chem vol. 47 (1975) 1730 (which is fully incorporated herein by reference), a Taundsen glow discharge was proposed for ionizing dopant gas in a CI source. U.S. Pat. No. 4,321,467 (which is fully incorporated herein by reference) proposes organics ionization in flow-afterglow at mbar-level gas pressures. VG Analytics introduced liquid samples via Thermospray interface and induced a glow discharge in the fore-vacuum region, as described in U.S. Pat. Nos. 4,647,772 and 4,794,252 (each of which is fully incorporated herein by reference). In U.S. Pat. No. 4,849,628 (which is fully incorporated herein by reference), McLuckey suggested sampling of liquid vapors from atmospheric pressure region into a glow discharge within a fore-vacuum stage at one to ten mbar gas pressure. Lubman et al. suggested ionized gaseous and liquid samples within Helium glow discharge at atmospheric pressure, as described in Applied Spectroscopy, 44 (1990) 1391 and Anal. Chem. 64 (1992) 1426 (each of which is fully incorporated herein by reference). Numerous groups have attempted to improve the softness and analytical merits of the glow discharge ionization sources. In spite of large variety of glow discharge sources, the employed ionization methods are split between two categories: (a) direct ionization and (b) chemical ionization.
Direct ionization in glow discharges occurs primarily due to Penning ionization by excited metastable of noble gases, while minor channels correspond to charge transfer from discharge ions and to electron impact ionization. Such ionization is likely to be quantitative, but harsh. As an example, Bertand et. al in JASMS, 5 (1994) 305 (which is fully incorporated herein by reference), exposed organic analytes to mBar glow discharge and demonstrated linear signal response within five orders of dynamic range, while obtaining spectra with softness varying from EI to CI spectra. Both sensitivity and softness appear strongly dependent on the analyzed compound and on the parameters of ion source. Adding dopant gases improves the intensity of molecular protonated ions and forms spectra similar to CI ones, as shown by Mason et al. in Int. J. Mass Spectrom. Ion Proc. 91 (1989) 209 (which is fully incorporated herein by reference).
The chemical ionization in glow discharges (or by sampled products of glow discharge) occurs primarily due to proton transfer from protonated water clusters, originating from ubiquitous water traces in technical purity gases. Proton transfer from water clusters has been intentionally promoted in a controlled proton transfer reaction (PTR) mass spectrometry as described by Hansel et al. in Int. J. Mass Spectrom. Ion Proc. 149 (1995) 609 (which is fully incorporated herein by reference). Lubman et al. ionized gaseous and liquid samples within Helium glow discharge at atmospheric pressure, as described in Applied Spectroscopy, 44 (1990) 1391 and Anal. Chem. 64 (1992) 1426 (each of which is fully incorporated herein by reference). Adding water with liquid samples substantially improved softness of organic spectra. However, the proton transfer reactions caused non-linear signal per concentration response and non-uniform ionization. Efficiency of ionization varied within three orders of magnitude between analyzed compounds. Proton affinity is known to depend on compound polarity, which explains the non-uniform ionization between chemical classes. Operation at atmospheric (as compared to mbar) pressures increases the role of ion molecular reactions, which explains mutual analyte interferences and matrix suppression effects, even at large excess of the charging agent.
A DART glow discharge method has been described in Andrade et. al, Anal. Chem, 80 (2008) 2646-2653 (which is fully incorporated herein by reference), where volatile compounds are mixed with glow discharge in helium at atmospheric pressure. Though the method describes Penning ionization as the main mechanism, large number of gas collisions lead to significant distortions by ion molecular reactions. Thus, this also produces protonated ions for polar compounds and, hence, is also prone to discrimination and interference effects. Thus, glow discharge ionization is likely to be either (a) quantitative but harsh at direct GD ionization, or (b) soft but not quantitative at chemical ionization, primary implemented by proton transfer from water clusters.
In WO 2012/024570 (which is fully incorporated herein by reference), the inventors of this disclosure attempted to soften direct ionization in glow discharge by using a conditioner (i.e. a conductive tube for controlling plasma residence time prior to sampling discharge products into an ion-molecular reactor with analyte). GC inlet, purified gases, and clean materials were used to reduce the amount of quenching parasitic vapors. However, the conditioning appears strongly dependent on trace amount of vapors, strongly reduces efficiency of ionization, so the choice remained the same—either quantitative or soft.